CN105550804A

CN105550804A - Machine tool product manufacturing system energy efficiency evaluation method based on gray fuzzy algorithm

Info

Publication number: CN105550804A
Application number: CN201510902387.4A
Authority: CN
Inventors: 谢广喜; 王艳; 毛志慧; 高云; 纪志成
Original assignee: Jiangnan University
Current assignee: Jiangnan University
Priority date: 2015-12-09
Filing date: 2015-12-09
Publication date: 2016-05-04

Abstract

The present invention proposes a method for evaluating the energy efficiency of the machine tool manufacturing system based on the gray fuzzy algorithm. It first establishes a comprehensive evaluation index system for the energy efficiency of the machine tool manufacturing system, and then uses the idea of combination weighting to combine the rough set theory with the analytic hierarchy process. To determine the weight of each index of energy efficiency of the machine tool product manufacturing system, the combination of qualitative analysis and quantitative analysis is realized, and the determination of the weight of the energy efficiency evaluation index of the machine tool product manufacturing system is more scientific and reasonable. On this basis, the present invention integrates the gray relational theory and the triangular membership model, and proposes an improved gray fuzzy energy efficiency analysis method, which better overcomes the subjectivity and objectivity in the energy efficiency evaluation process. This method is scientific and reasonable. By calculating energy efficiency, enterprises can improve energy efficiency in a targeted manner and promote green manufacturing.

Description

Energy Efficiency Evaluation Method of Machine Tool Product Manufacturing System Based on Gray Fuzzy Algorithm

技术领域technical field

本发明涉及一种机床产品制造系统能效评价方法，涉及机床产品综合评价、绿色生产研究领域。The invention relates to an energy efficiency evaluation method for a machine tool product manufacturing system, and relates to the fields of comprehensive evaluation of machine tool products and green production research.

背景技术Background technique

制造业作为国民经济的支柱产业，在创造巨大经济财富的同时，也消耗了大量制造资源特别是能源，并造成了对环境的严重影响。能源问题已成为制约经济和社会发展的直观因素，从对能源利用的方向出发，节能成为重中之重。典型的机床制造系统的基本构成要素可以分为生产环境、生产设备、生产对象、操作者四个部分。制造系统在生产过程中消耗的能量可以分为直接能量和间接能量，直接能量是制造产品的各种过程消耗的能量，间接能量是为了维护车间内的生产环境需要消耗的能量。As a pillar industry of the national economy, manufacturing has created huge economic wealth, but it also consumes a lot of manufacturing resources, especially energy, and has caused serious impacts on the environment. Energy issues have become an intuitive factor restricting economic and social development. Starting from the direction of energy utilization, energy conservation has become a top priority. The basic components of a typical machine tool manufacturing system can be divided into four parts: production environment, production equipment, production objects, and operators. The energy consumed by the manufacturing system in the production process can be divided into direct energy and indirect energy. Direct energy is the energy consumed by various processes of manufacturing products, and indirect energy is the energy that needs to be consumed to maintain the production environment in the workshop.

加强企业能效评价、提高系统制造系统能量效率已成为制造业的当务之急。能效评价，即对企业在生产过程中的能效利用情况进行评价，促使企业改进生产工艺和管理方式，从而有利用提高能源利用效率，节约能源。制造系统能效评价包括制造系统能量消耗状态及能量消耗过程的分析评价以及在此基础上对能量效率的评价。提高能源利用效率的前提是了解用能系统本身的用能情况，因此研究能效测评方法，建立完善的能效评估指标体系具有现实意义。Strengthening the energy efficiency evaluation of enterprises and improving the energy efficiency of system manufacturing systems has become a top priority for the manufacturing industry. Energy efficiency evaluation, that is, to evaluate the energy efficiency utilization of enterprises in the production process, to urge enterprises to improve production processes and management methods, so as to effectively improve energy utilization efficiency and save energy. The energy efficiency evaluation of manufacturing system includes the analysis and evaluation of the energy consumption state and energy consumption process of the manufacturing system and the evaluation of energy efficiency on this basis. The premise of improving energy utilization efficiency is to understand the energy consumption of the energy-using system itself, so it is of practical significance to study energy efficiency evaluation methods and establish a complete energy efficiency evaluation index system.

发明内容Contents of the invention

本发明的目的在于提供一种机床产品制造系统能效评价方法,该方法避免了专家主观因素的影响,同时也避免了当样本数据不够全面的情况下，所获得的权重将严重偏离现实的问题，能为机床产品综合评价提供依据和指导。The purpose of the present invention is to provide a method for evaluating the energy efficiency of a machine tool product manufacturing system, which avoids the influence of subjective factors of experts, and also avoids the problem that the obtained weight will seriously deviate from reality when the sample data is not comprehensive enough. It can provide the basis and guidance for the comprehensive evaluation of machine tool products.

为了达到上述目的,本发明所述的机床产品制造系统能效评价方法，包括如下步骤：In order to achieve the above object, the energy efficiency evaluation method of the machine tool product manufacturing system according to the present invention comprises the following steps:

步骤一、建立机床产品制造系统能效评价指标体系，能效评价指标体系中所有具体指标构成评价因素集C；Step 1. Establish the energy efficiency evaluation index system of the machine tool product manufacturing system, and all the specific indicators in the energy efficiency evaluation index system constitute the evaluation factor set C;

步骤二、应用粗糙集和层次分析法的组合方法确定指标的权重集合W；即利用粗糙集与层次分析法分别获得客观、主观两个方面的指标权重值，对两者进行综合，获得最后指标权重值，得到一组最终的评价指标权重Step 2. Apply the combined method of rough set and AHP to determine the weight set W of the index; that is, use the rough set and AHP to obtain the objective and subjective index weight values respectively, and synthesize the two to obtain the final index Weight value to get a set of final evaluation index weights

W＝μw_Ai+(1-μ)w_Bi W＝μw _Ai +(1-μ)w _Bi

其中w_Ai是指客观权重值，w_Bi是指主观权重值，μ∈[0,1]，μ的取值根据具体情况而定，μ越接近于0表示决策越倾向于专家经验，μ越接近于1表示决策越倾向于客观数据；Among them, w _Ai refers to the objective weight value, w _Bi refers to the subjective weight value, μ ∈ [0,1], and the value of μ depends on the specific situation. The closer μ is to 0, the more inclined the decision-making is to expert experience. Closer to 1 means that the decision is more inclined to objective data;

步骤三、应用线性比例变换的方法对机床产品制造系统原始定量指标数据进行无量纲化处理；Step 3, applying the method of linear proportional transformation to carry out dimensionless processing on the original quantitative index data of the machine tool product manufacturing system;

步骤四、应用分级打分法对机床产品制造系统原始定性指标数据进行定量化处理；Step 4: Quantitatively process the original qualitative index data of the machine tool product manufacturing system by applying the grading and scoring method;

步骤五、应用三角形隶属模型确定单因素模糊评价集；Step five, applying the triangular membership model to determine the single-factor fuzzy evaluation set;

步骤六、根据灰色关联法计算出一级指标评价矩阵，进而得到一级指标评价结果；Step 6. Calculate the first-level index evaluation matrix according to the gray relational method, and then obtain the first-level index evaluation results;

步骤七、利用灰色关联法综合交互评价出多层指标。Step 7: Use the gray relational method to comprehensively and interactively evaluate the multi-layer indicators.

具体的，步骤一所述能效评价指标体系包括经济能效指标、产品能效指标、设备能效指标和任务流程能效指标4个一级指标，所述经济能效指标包括的二级指标有：万元产品能耗、万元增加值能耗，所述产品能效指标包括的二级指标有：单位产品综合能耗、单位产品节能量、产品用能水平，所述设备能效指标包括的二级指标有：机床设备能效、能源输送效率、能源加工转换效率，所述任务流程能效指标包括的二级指标有：生产工艺能效、生产资源调度能效，这10个二级指标构成评价因素集C。Specifically, the energy efficiency evaluation index system described in step 1 includes four first-level indicators: economic energy efficiency indicators, product energy efficiency indicators, equipment energy efficiency indicators, and task process energy efficiency indicators. The second-level indicators included in the economic energy efficiency indicators are: energy consumption, energy consumption per 10,000 yuan added value, the secondary indicators included in the product energy efficiency indicators include: comprehensive energy consumption per unit product, energy saving energy per unit product, product energy consumption level, the secondary indicators included in the equipment energy efficiency indicators include: machine tool Equipment energy efficiency, energy transmission efficiency, energy processing conversion efficiency, the secondary indicators included in the task process energy efficiency indicators include: production process energy efficiency, production resource scheduling energy efficiency, these 10 secondary indicators constitute evaluation factor set C.

步骤三中，设第k个指标的原始数据值为则要经过下式进行无量纲化处理，其中处理后的数据值C_i(k)∈(0,1),In step 3, set the original data value of the k-th index as Then it needs to go through the following formula for dimensionless processing, where the processed data value C _i (k)∈(0,1),

${C C}_{i i} ((k k)) = = \frac{{c c}_{k k}^{i i} - - min min {c c}_{k k}^{i i}}{max max {c c}_{k k}^{i i} - - min min {c c}_{k k}^{i i}}$

且i＝1,2…n,k＝1,2…m，其中m为决策指标数量，n为可选方案数量。And i=1,2...n, k=1,2...m, where m is the number of decision-making indicators, and n is the number of optional solutions.

步骤四把定性指标转化为定量指标，采用分级打分法，对每级赋予一个分值。Step 4 converts qualitative indicators into quantitative indicators, adopts the graded scoring method, and assigns a score to each grade.

步骤五从单个指标出发，确定评价集元素的隶属度；从U到F(V)的模糊映射:Step 5 Starting from a single index, determine the membership degree of the evaluation set elements; fuzzy mapping from U to F(V):

$f f : : U u &RightArrow; &Right Arrow; F f ((V V)),, &ForAll; &ForAll; {u u}_{i i} &Element; &Element; U u,, {u u}_{i i} | | &RightArrow; &Right Arrow; f f (({u u}_{i i})) = = \frac{{r r}_{i i,, 11}}{{c c}_{11}} + + \frac{{r r}_{i i,, 22}}{{c c}_{22}} + + ... ... + + \frac{{r r}_{i i,, k k}}{{c c}_{k k}} ... ... + + \frac{{r r}_{i i,, m m}}{{c c}_{m m}}$

式中,r_i,k表示u_i属于c_k的隶属度。In the formula, ri _,k represents the degree of membership that u _i belongs to c _k .

步骤六根据灰色关联法实现一级指标的综合评价，最优指标集为： $C * = [\begin{matrix} c_{1}^{*} & c_{2}^{*} & ... & c_{m}^{*} \end{matrix}],$ 原始评价矩阵为： $D = [\begin{matrix} c_{1}^{*} & c_{2}^{*} & ... & c_{m}^{*} \\ c_{1}^{1} & c_{2}^{1} & ... & c_{m}^{1} \\ ... & ... & ... & ... \\ c_{1}^{n} & c_{2}^{n} & ... & c_{m}^{n} \end{matrix}]$ Step 6 Realize the comprehensive evaluation of the first-level indicators according to the gray relational method, and the optimal index set is: $C * = [\begin{matrix} c_{1}^{*} & c_{2}^{*} & ... & c_{m}^{*} \end{matrix}],$ The original evaluation matrix is: $D. = [\begin{matrix} c_{1}^{*} & c_{2}^{*} & ... & c_{m}^{*} \\ c_{1}^{1} & c_{2}^{1} & ... & c_{m}^{1} \\ ... & ... & ... & ... \\ c_{1}^{no} & c_{2}^{no} & ... & c_{m}^{no} \end{matrix}]$

式中，m为决策指标数量，n为可选方案数量，为第k个指标的最优值，为第i个方案中第k个指标的原始值；可得出两极最小差： In the formula, m is the number of decision-making indicators, n is the number of alternatives, is the optimal value of the kth index, is the original value of the k-th indicator in the i-th scheme; the minimum difference between the two poles can be obtained:

两极最大差： ${TOW}_{m a x} = \underset{i}{m a x} \underset{k}{m a x} | c_{k}^{*} - c_{k}^{i} |$ Maximum difference between two poles: ${TOW}_{m a x} = \underset{i}{m a x} \underset{k}{m a x} | c_{k}^{*} - c_{k}^{i} |$

灰色关联系数为：The gray correlation coefficient is:

${L L}_{i i}^{k k} = = \frac{{TOW TOW}_{m m i i n no} + + {ρTOW ρTOW}_{m m a a x x}}{| | {c c}_{k k}^{* *} - - {c c}_{k k}^{i i} | | + + {ρTOW ρTOW}_{m m a a x x}},, ρ ρ &Element; &Element; ((00,, 11))$

评价矩阵为：The evaluation matrix is:

$R R = = [\begin{matrix} {L L}_{11} ((11)) & {L L}_{22} ((11)) & ... ... & {L L}_{n no} ((11)) \\ {L L}_{11} ((22)) & {L L}_{22} ((22)) & ... ... & {L L}_{n no} ((22)) \\ ... ... & ... ... & ... ... & ... ... \\ {L L}_{11} ((m m)) & {L L}_{22} ((m m)) & ... ... & {L L}_{n no} ((m m)) \end{matrix}]$

最后灰色综合评价：The final gray comprehensive evaluation:

J＝W×RJ=W×R

式中，W为权重矩阵，R为评价矩阵。In the formula, W is the weight matrix, and R is the evaluation matrix.

步骤七实现多级灰色综合评价：若指标有y层，则要进行y级灰色综合评价,c_k作为第k个评价指标，它的单指标评价集其中s作为指标数量；当指标有两层且每层有多个指标时，先对第二层指标进行单指标模糊评价，再由第二层指标对第一层指标进行一级灰色综合评价，再由第一层指标的一级灰色综合评价结果对第二层指标进行二级灰色综合评价，评价结果即为系统评价结果。Step 7: Realize multi-level gray comprehensive evaluation: if the index has y layers, it needs to carry out y-level gray comprehensive evaluation, c _k is used as the kth evaluation index, and its single index evaluation set Among them, s is the number of indicators; when the indicators have two layers and each layer has multiple indicators, the single-index fuzzy evaluation of the second-layer indicators is performed first, and then the first-level gray comprehensive evaluation is performed on the first-layer indicators by the second-layer indicators. Then, the first-level gray comprehensive evaluation results of the first-level indicators are used for the second-level gray comprehensive evaluation of the second-level indicators, and the evaluation results are the system evaluation results.

本发明的有益效果是：本发明首先建立了机床产品制造系统能效综合评价指标体系，然后采用组合赋权的思想，将粗糙集理论与层次分析法结合来确定机床产品制造系统能效各指标的权重,实现了定性分析与定量分析的结合,使机床产品制造系统能效评价指标权重的确定更科学、更合理。在此基础上,本发明综合灰色关联理论和三角形隶属模型，提出了一种改进的灰色模糊能效分析方法，较好地克服了能效评价过程中的主观性和客观性。该方法是科学合理的，企业通过计算能效，可以针对性地进行能效改进，推进绿色制造。The beneficial effects of the present invention are: firstly, the present invention establishes the energy efficiency comprehensive evaluation index system of the machine tool product manufacturing system, and then adopts the idea of combination weighting to determine the weight of each index of the machine tool product manufacturing system energy efficiency by combining the rough set theory and the analytic hierarchy process , realizes the combination of qualitative analysis and quantitative analysis, and makes the determination of the weight of the energy efficiency evaluation index of the machine tool product manufacturing system more scientific and reasonable. On this basis, the present invention integrates the gray relational theory and the triangular membership model, and proposes an improved gray fuzzy energy efficiency analysis method, which better overcomes the subjectivity and objectivity in the energy efficiency evaluation process. This method is scientific and reasonable. By calculating energy efficiency, enterprises can improve energy efficiency in a targeted manner and promote green manufacturing.

附图说明Description of drawings

图1是本发明的能效评价流程图。Fig. 1 is a flow chart of the energy efficiency evaluation of the present invention.

图2是本发明的综合评价指标体系。Fig. 2 is the comprehensive evaluation index system of the present invention.

具体实施方式detailed description

本发明主要是针对机床产品制造系统能效综合评价提供了一种评价方法，如图1所示，该方法主要包括以下几个步骤：步骤一、建立机床产品制造系统能效评价指标体系和评价因素集C；步骤二、应用粗糙集—AHM(层次分析法)组合方法确定指标的权重集合W；步骤三、应用线性比例变换的方法对机床产品制造系统原始定量指标数据进行无量纲化处理；步骤四、应用分级打分法对机床产品制造系统原始定性指标数据进行定量化处理。步骤五、应用三角形隶属模型确定单因素模糊评价集；步骤六、根据灰色关联法计算出一级评价矩阵，进而得到一级评价结果；步骤七、利用灰色关联法综合交互评价出多层指标。The present invention mainly provides an evaluation method for the comprehensive evaluation of the energy efficiency of the machine tool product manufacturing system, as shown in Figure 1, the method mainly includes the following steps: Step 1, establishing the energy efficiency evaluation index system and evaluation factor set of the machine tool product manufacturing system C; Step 2, applying the rough set-AHM (Analytic Hierarchy Process) combination method to determine the weight set W of the index; Step 3, applying the method of linear proportional transformation to carry out dimensionless processing on the original quantitative index data of the machine tool product manufacturing system; Step 4 1. Quantitatively process the original qualitative index data of the machine tool product manufacturing system by applying the grading and scoring method. Step 5: Use the triangular membership model to determine the single-factor fuzzy evaluation set; Step 6: Calculate the first-level evaluation matrix according to the gray relational method, and then obtain the first-level evaluation results; Step 7: Use the gray relational method to comprehensively and interactively evaluate multi-layer indicators.

步骤一中：评价指标的选取必须注意评价的目的性、全面性、稳定性与可行性原则，评价指标的确定要以实际情况为基础，这里选取经济能效、产品能效、设备能效和任务流程能效4个一级指标和10个二级指标建立能源评价指标体系，全面涵盖了机床产品制造系统、产品层、设备层和任务层各指标，而传统的机床产品制造系统能效评价体系，忽略了生产工艺能效和生产资源调度能效。本发明的整个指标体系的层次结构如如图2所示。该能效评价指标体系包括经济能效指标B1、产品能效指标B2、设备能效指标B3和任务流程能效指标B4共4个一级指标，所述经济能效指标B1包括的二级指标有：万元产品能耗C1、万元增加值能耗C2，所述产品能效指标B2包括的二级指标有：单位产品综合能耗C3、单位产品节能量C4、产品用能水平C5，所述设备能效指标B3包括的二级指标有：机床设备能效C6、能源输送效率C7、能源加工转换效率C8，所述任务流程能效指标B4包括的二级指标有：生产工艺能效C9、生产资源调度能效C10，这10个二级指标构成评价因素集C。Step 1: The selection of evaluation indicators must pay attention to the principles of purpose, comprehensiveness, stability and feasibility of the evaluation. The determination of evaluation indicators should be based on the actual situation. Here, economic energy efficiency, product energy efficiency, equipment energy efficiency and task process energy efficiency are selected. 4 first-level indicators and 10 second-level indicators establish an energy evaluation index system that comprehensively covers the indicators of the machine tool product manufacturing system, product layer, equipment layer, and task layer, while the traditional energy efficiency evaluation system of machine tool product manufacturing systems ignores production. Process energy efficiency and production resource scheduling energy efficiency. The hierarchical structure of the entire index system of the present invention is shown in FIG. 2 . The energy efficiency evaluation index system includes four first-level indicators: economic energy efficiency index B1, product energy efficiency index B2, equipment energy efficiency index B3, and task process energy efficiency index B4. The economic energy efficiency index B1 includes two secondary indicators: Consumption C1, energy consumption of 10,000 yuan added value C2, the secondary indicators included in the product energy efficiency index B2 are: comprehensive energy consumption per unit product C3, energy saving per unit product C4, product energy consumption level C5, the equipment energy efficiency index B3 includes The secondary indicators include: machine tool equipment energy efficiency C6, energy transmission efficiency C7, energy processing conversion efficiency C8, the task process energy efficiency indicator B4 includes the following secondary indicators: production process energy efficiency C9, production resource scheduling energy efficiency C10, these 10 The secondary indicators constitute the evaluation factor set C.

步骤二利用粗糙集理论在处理不确定、不精确数据的优势，能够获得较为客观的指标权重信息；另一方面利用AHM能够充分利用领域专家经验的优点，获得专家对指标客观的重要性评价结果，克服传统层次分析法在评价时对一致性检验要求较高的不足。The second step is to use the advantages of rough set theory in dealing with uncertain and imprecise data to obtain more objective index weight information; on the other hand, to use AHM to make full use of the advantages of domain expert experience to obtain the objective evaluation results of the importance of indicators by experts , to overcome the shortcomings of the traditional AHP in the evaluation of the consistency test requirements.

(1)基于粗糙集理论的权重计算方法：(1) Weight calculation method based on rough set theory:

在决策表S＝(U,C,D,V,f)中，决策属性D(U/D＝{D₁,D₂,...D_k})相对于条件属性集C(U/C＝{C₁,C₂,...C_m})的条件信息熵为：In the decision table S=(U,C,D,V,f), the decision attribute D(U/D={D ₁ ,D ₂ ,...D _k }) is relative to the condition attribute set C(U/C ={C ₁ ,C ₂ ,...C _m }) the conditional information entropy is:

$I I ((D D. | | C C)) = = {Σ Σ}_{i i = = 11}^{m m} \frac{{| | C C | |}^{22}}{{| | U u | |}^{22}} {Σ Σ}_{j j = = 11}^{k k} \frac{{| | {D D.}_{j j} \cap \cap {C C}_{i i} | |}^{22}}{{| | {C C}_{i i} | |}^{22}} ((11 - - \frac{| | {D D.}_{j j} \cap \cap {C C}_{i i} | |}{| | {C C}_{i i} | |}))$

其中U是对象集合，子集C是条件属性集，D是决策属性集，C∩D＝φ，D≠φ,V是属性值集合，f代表一个信息函数，它表示论域中每一个对象在相应属性上所取到的属性值。Among them, U is the object set, the subset C is the condition attribute set, D is the decision attribute set, C∩D=φ, D≠φ, V is the attribute value set, f represents an information function, which represents each object in the domain of discourse The attribute value obtained on the corresponding attribute.

在决策表S＝(U,C,D,V,f)中，则条件属性(指标)C的重要度定义为In the decision table S=(U,C,D,V,f), Then the importance of the condition attribute (index) C is defined as

Sig(c)＝I(D|C-{c}-I(D|C))Sig(c)=I(D|C-{c}-I(D|C))

在决策表S＝(U,C,D,V,f)中，则条件属性(指标)C的权重为In the decision table S=(U,C,D,V,f), Then the weight of the condition attribute (index) C is

${W W}_{A A i i} ((c c)) = = \frac{S S i i g g ((c c)) + + I I ((D D. | | {{c c}}))}{\underset{a a &Element; &Element; C C}{Σ Σ} {{S S i i g g ((a a)) + + I I ((D D. | | {{a a}}}}}$

(2)基于AHM的权重计算方法：(2) AHM-based weight calculation method:

为计算同层元素之间的相对重要性，建立判断矩阵A＝{a_ij}，其中a_ij＝1/a_ji，a_ii＝1。其中a_ij是根据专家知识得到的重要度参数，a_ij∈{1,3,5,7,9}。将A＝{a_ij}通过公式转化为测度矩阵In order to calculate the relative importance between elements in the same layer, a judgment matrix A={a _ij } is established, where a _ij =1/a _ji , a _ii =1. Where a _ij is the importance parameter obtained according to expert knowledge, a _ij ∈ {1,3,5,7,9}. Convert A={a _ij } into a measure matrix through the formula

$μ μ = = \{\begin{matrix} \frac{β β k k}{β β k k + + 11} & {a a}_{i i j j} = = k k \\ \frac{11}{β β k k + + 11} & {a a}_{i i j j} = = \frac{11}{k k} \\ 0.5 0.5 & {a a}_{i i j j} = = 11,, i i &NotEqual; &NotEqual; j j \\ 00 & {a a}_{i i j j} = = 11,, i i = = j j \end{matrix}$

K为大于1的正整数，取β＝1K is a positive integer greater than 1, take β=1

计算单层指标权重，得到每层指标相对于其上层指标的加权子集:Calculate the weight of single-layer indicators to obtain the weighted subset of each layer indicator relative to its upper layer indicators:

W＝[w₁,w₂...w₁₀]，W=[w ₁ ,w ₂ . . . w ₁₀ ],

${w w}_{i i} = = \frac{22}{n no ((n no - - 11))} {Σ Σ}_{j j = = 11}^{n no} {μ μ}_{i i j j},, i i = = 11,, 22,, ... ...,, n no,,$

${Σ Σ}_{i i = = 11}^{n no} {w w}_{i i} = = 11,, 00 \leq \leq {w w}_{i i} \leq \leq 11,,$

n＝10。n=10.

计算底层元素的组合权重Calculate the combined weight of the underlying elements

w_j＝w_i*w_ij w _j =w _i *w _ij

其中w_j表示第j项子目标对于总目标的组合权重，w_i表示第i项子目标的组合权重，w_ij表示第j项子目标对第i项子目标的权重，其中第i项子目标位于第j项子目标的上一层。组合权重主要用来分析各个指标的之间的重要性，不用于后面的计算。Among them, w _j represents the combined weight of the j-th sub-goal to the total target, w _i represents the combined weight of the i-th sub-goal, and w _ij represents the weight of the j-th sub-goal to the i-th sub-goal, where the i-th sub-goal The goal is located one level above the jth subgoal. Combination weights are mainly used to analyze the importance of each indicator, and are not used for subsequent calculations.

(3)评价指标综合权重计算函数构建：(3) Construction of evaluation index comprehensive weight calculation function:

利用粗糙集与AHM方法分别获得客观、主观两个方面的指标权重值，利用粗糙集理论可以处理不确定、不精确的数据,克服了受专家主观因素的影响，却也容易受样本数据的选择影响，特别是在样本数据不够全面的情况下，所获得的权重将严重偏离现实。而AHM方法可以充分利用专家的经验，但是受人为影响很大，不能客观反映样本权重。因此对两者进行综合，获得最后指标权重值，得到一组最终的评价指标权重。Using rough set and AHM methods to obtain objective and subjective index weight values respectively, using rough set theory to deal with uncertain and inaccurate data, overcoming the influence of subjective factors of experts, but also easily affected by the selection of sample data Influence, especially when the sample data is not comprehensive enough, the obtained weights will seriously deviate from reality. The AHM method can make full use of the experience of experts, but it is greatly influenced by human beings and cannot objectively reflect the weight of samples. Therefore, the two are integrated to obtain the final index weight value, and a set of final evaluation index weights is obtained.

W＝μw_Ai+(1-μ)w_Bi W＝μw _Ai +(1-μ)w _Bi

其中w_Ai是指客观权重值，w_Bi是指主观权重值，μ的取值根据具体情况而定，当决策倾向专家经验时，μ∈[0,0.5)，而当决策倾向客观数据时，μ∈(0.5,1]。最后计算所得到的权重，即为由主观和客观权重综合计算所得到的最后指标评价中的权重。Among them, w _Ai refers to the objective weight value, w _Bi refers to the subjective weight value, and the value of μ depends on the specific situation. When the decision is inclined to expert experience, μ∈[0,0.5), and when the decision is inclined to objective data, μ∈(0.5,1]. The weight obtained by the final calculation is the weight in the final index evaluation obtained by comprehensive calculation of subjective and objective weights.

步骤三实现了定量指标的无量纲化处理，对于定量指标而言，由于各指标的计量单位、量级不同，须对原始数据指标进行无量纲化处理，以减少随机因素的干扰。设第k个指标的原始数据值为则要经过下式进行无量纲化处理，其中处理后的数据值C_i(k)∈(0,1)。Step 3 realizes the dimensionless processing of quantitative indicators. For quantitative indicators, due to the different measurement units and magnitudes of each indicator, it is necessary to perform dimensionless processing on the original data indicators to reduce the interference of random factors. Let the original data value of the kth indicator be Then, the dimensionless processing needs to be performed through the following formula, wherein the processed data value C _i (k)∈(0,1).

步骤四把定性指标转化为定量指标，本文采用分级打分法，对每级赋予一个分值，若等级为“优、良、中、差”，则分值分别为“4、3、2、1”。Step 4 converts qualitative indicators into quantitative indicators. In this paper, a graded scoring method is used to assign a score to each level. If the grade is "excellent, good, medium, and poor", the scores are "4, 3, 2, 1" respectively. ".

步骤五从单个指标出发，确定评价集元素的隶属度。从U到F(V)的模糊映射:Step five starts from a single indicator to determine the membership degree of the evaluation set elements. Fuzzy mapping from U to F(V):

$f f : : U u &RightArrow; &Right Arrow; F f ((V V)),, &ForAll; &ForAll; {u u}_{i i} &Element; &Element; U u,, {u u}_{i i} | | &RightArrow; &Right Arrow; f f (({u u}_{i i})) = = \frac{{r r}_{i i,, 11}}{{c c}_{11}} + + \frac{{r r}_{i i,, 22}}{{c c}_{22}} + + ... ... + + \frac{{r r}_{i i,, m m}}{{c c}_{m m}}$

式中,r_i,j表示u_i属于c_j的隶属度。In the formula, r _{i, j} represents the membership degree of u _i belonging to c _j .

确定隶属函数的方法有函数推理法、二元对比排序法、模糊统计法、三分法以及模糊分布法等。本文采用三角形隶属模型，隶属函数如下:The methods to determine the membership function include function inference method, binary comparison and sorting method, fuzzy statistical method, third-point method and fuzzy distribution method. In this paper, the triangular membership model is adopted, and the membership function is as follows:

适用于值越小越好的指标x，其隶属函数模型如下：It is suitable for the index x whose value is smaller and better, and its membership function model is as follows:

${u u}_{11} ((x x)) = = \{\begin{matrix} 11 & x x \leq \leq {a a}_{11} \\ ((x x - - {a a}_{22})) / / (({a a}_{11} - - {a a}_{22})) & {a a}_{11} < < x x \leq \leq {a a}_{22} \\ 00 & x x > > {a a}_{22} \end{matrix}$

${u u}_{22} ((x x)) = = \{\begin{matrix} 00 & x x \leq \leq {a a}_{11} o o r r x x &GreaterEqual; &Greater Equal; {a a}_{33} \\ ((x x - - {a a}_{11})) / / (({a a}_{22} - - {a a}_{11})) & {a a}_{11} < < x x < < {a a}_{22} \\ ((x x - - {a a}_{33})) (({a a}_{22} - - {a a}_{33})) & {a a}_{22} < < x x < < {a a}_{33} \end{matrix}$

${u u}_{33} ((x x)) = = \{\begin{matrix} 00 & x x \leq \leq {a a}_{22} \\ ((x x - - {a a}_{22})) / / (({a a}_{33} - - {a a}_{22})) & {a a}_{22} < < x x \leq \leq {a a}_{33} \\ 11 & x x > > {a a}_{33} \end{matrix}$

适用于值越大越好的指标x，其隶属函数模型如下：It is suitable for the index x whose value is better as it is larger, and its membership function model is as follows:

${u u}_{11} ((x x)) = = \{\begin{matrix} 00 & x x \leq \leq {a a}_{22} \\ ((x x - - {a a}_{22})) / / (({a a}_{11} - - {a a}_{22})) & {a a}_{22} < < x x \leq \leq {a a}_{11} \\ 11 & x x > > {a a}_{11} \end{matrix}$

${u u}_{22} ((x x)) = = \{\begin{matrix} 00 & x x \leq \leq {a a}_{33} o o r r x x &GreaterEqual; &Greater Equal; {a a}_{11} \\ ((x x - - {a a}_{33})) / / (({a a}_{22} - - {a a}_{33})) & {a a}_{33} < < x x \leq \leq {a a}_{22} \\ ((x x - - {a a}_{33})) (({a a}_{22} - - {a a}_{33})) & {a a}_{22} < < x x < < {a a}_{11} \end{matrix}$

${u u}_{33} ((x x)) = = \{\begin{matrix} 11 & x x \leq \leq {a a}_{33} \\ ((x x - - {a a}_{22})) / / (({a a}_{33} - - {a a}_{22})) & {a a}_{33} < < x x \leq \leq {a a}_{22} \\ 00 & x x > > {a a}_{22} \end{matrix}$

适用于值应在某固定区间的指标x，其隶属函数模型如下:Applicable to the index x whose value should be in a fixed interval, its membership function model is as follows:

${u u}_{11} ((x x)) = = \{\begin{matrix} 11 & {a a}_{1111} \leq \leq x x \leq \leq {a a}_{1212} \\ ((x x - - {a a}_{1111})) (({a a}_{1111} - - {a a}_{21 twenty one})) & {a a}_{21 twenty one} \leq \leq x x < < {a a}_{1111} \\ ((x x - - {a a}_{22 twenty two})) (({a a}_{1212} - - {a a}_{22 twenty two})) & {a a}_{1212} < < x x < < {a a}_{22 twenty two} \\ 00 & x x < < {a a}_{21 twenty one} o o r r x x > > {a a}_{22 twenty two} \end{matrix}$

${u u}_{22} ((x x)) = = \{\begin{matrix} 00 & x x \leq \leq {a a}_{3131} o o r r x x &GreaterEqual; &Greater Equal; {a a}_{3232} o o r r {a a}_{1111} \leq \leq x x \leq \leq {a a}_{1212} \\ ((x x - - {a a}_{3131})) / / (({a a}_{21 twenty one} - - {a a}_{3131})) & {a a}_{3131} < < x x < < {a a}_{1111} \\ ((x x - - {a a}_{1111})) / / (({a a}_{21 twenty one} - - {a a}_{1111})) & {a a}_{21 twenty one} \leq \leq x x < < {a a}_{1111} \\ ((x x - - {a a}_{1212})) / / (({a a}_{22 twenty two} - - {a a}_{1212})) & {a a}_{1212} \leq \leq x x < < {a a}_{22 twenty two} \\ ((x x - - {a a}_{3232})) / / (({a a}_{22 twenty two} - - {a a}_{3232})) & {a a}_{22 twenty two} \leq \leq x x < < {a a}_{3232} \end{matrix}$

${u u}_{33} ((x x)) = = \{\begin{matrix} 11 & x x \leq \leq {a a}_{3131} o o r r x x &GreaterEqual; &Greater Equal; {a a}_{3232} \\ ((x x - - {a a}_{1111})) (({a a}_{1111} - - {a a}_{21 twenty one})) & {a a}_{3131} < < x x < < {a a}_{21 twenty one} \\ ((x x - - {a a}_{22 twenty two})) (({a a}_{1212} - - {a a}_{22 twenty two})) & {a a}_{22 twenty two} < < x x < < {a a}_{3232} \\ 00 & {a a}_{21 twenty one} \leq \leq x x \leq \leq {a a}_{22 twenty two} \end{matrix}$

u₁，u₂，u₃是表示单因素好中差的隶属度，且u₁，u₂，u₃满足如下关系：u ₁ , u ₂ , u ₃ are the membership degrees representing the single factor good, medium and poor, and u ₁ , u ₂ , u ₃ satisfy the following relationship:

u₁+u₂+u₃＝1u ₁ +u ₂ +u ₃ =1

由f(u_i)可得到单因素评价集：The single-factor evaluation set can be obtained from f(u _i ):

R_i＝(r_i1,r_i2,…r_im)R _i ＝(r _i1 ,r _i2 ,…r _im )

根据单因素评价集，得出定性的评价结果。According to the single factor evaluation set, the qualitative evaluation results are obtained.

步骤六根据灰色关联法实现了一级指标的综合评价。最优指标集为： $C * = [\begin{matrix} c_{1}^{*} & c_{2}^{*} & ... & c_{m}^{*} \end{matrix}]$ 原始评价矩阵为： $D = [\begin{matrix} c_{1}^{*} & c_{2}^{*} & ... & c_{m}^{*} \\ c_{1}^{1} & c_{2}^{1} & ... & c_{m}^{1} \\ ... & ... & ... & ... \\ c_{1}^{n} & c_{2}^{n} & ... & c_{m}^{n} \end{matrix}]$ Step 6 realizes the comprehensive evaluation of the first-level indicators according to the gray relational method. The optimal index set is: $C * = [\begin{matrix} c_{1}^{*} & c_{2}^{*} & ... & c_{m}^{*} \end{matrix}]$ The original evaluation matrix is: $D. = [\begin{matrix} c_{1}^{*} & c_{2}^{*} & ... & c_{m}^{*} \\ c_{1}^{1} & c_{2}^{1} & ... & c_{m}^{1} \\ ... & ... & ... & ... \\ c_{1}^{no} & c_{2}^{no} & ... & c_{m}^{no} \end{matrix}]$

式中，m为决策指标数量，n为可选方案数量，为第k个指标的最优值，为第i个方案中第k个指标的原始值。可得出两极最小差： In the formula, m is the number of decision-making indicators, n is the number of alternatives, is the optimal value of the kth index, is the original value of the k-th index in the i-th scheme. The minimum difference between the two poles can be obtained:

灰色关联系数为：The gray correlation coefficient is:

评价矩阵为：The evaluation matrix is:

最后灰色综合评价：The final gray comprehensive evaluation:

J＝W×RJ=W×R

步骤七实现多级灰色综合评价：若指标有y层，则要进行y级灰色综合评价,c_k作为第k个评价指标，它的单指标评价集其中s作为指标数量。如当指标有两层且每层有多个指标时，先对第二层指标进行单指标模糊评价，再由第二层指标对第一层指标进行一级灰色综合评价，再由第一层指标的一级灰色综合评价结果对第二层指标进行二级灰色综合评价，评价结果即为系统评价结果。Step 7: Realize multi-level gray comprehensive evaluation: if the index has y layers, it needs to carry out y-level gray comprehensive evaluation, c _k is used as the kth evaluation index, and its single index evaluation set where s is the number of indicators. For example, when there are two layers of indicators and each layer has multiple indicators, the single-index fuzzy evaluation of the second layer of indicators is performed first, and then the second layer of indicators performs a gray comprehensive evaluation of the first layer of indicators, and then the first layer of indicators The results of the first-level gray comprehensive evaluation of the indicators carry out the second-level gray comprehensive evaluation on the second-level indicators, and the evaluation results are the results of the system evaluation.

下面结合实施例对本发明作进一步描述。The present invention will be further described below in conjunction with embodiment.

选取甲、乙两家机床制造企业作为评价对象进行综合评价,各指标的具体数值如下表1所示。The two machine tool manufacturers A and B are selected as the evaluation objects for comprehensive evaluation. The specific values of each index are shown in Table 1 below.

表1各指标的数值Table 1 Values of each index

确定指标的权重集合WDetermine the weight set W of the indicator

计算基于粗糙集理论的权重:Calculate weights based on rough set theory:

根据公式(2)先计算各指标的重要度:Calculate the importance of each index according to the formula (2):

一级指标:First-level indicators:

$s the s i i g g (({B B}_{11})) = = \frac{99}{3434} s the s i i g g (({B B}_{22})) = = \frac{77}{3434} s the s i i g g (({B B}_{33})) = = \frac{88}{3434} s the s i i g g (({B B}_{44})) = = \frac{1010}{3434}$

二级指标:Secondary indicators:

sig(C₁)＝0.618sig(C₂)＝0.382sig(C₃)＝0.323sig(C₄)＝0.31sig(C₅)＝0.367sig(C₆)＝0.439sig(C ₁ )=0.618 sig(C ₂ )=0.382 sig(C ₃ )=0.323 sig(C ₄ )=0.31 sig(C ₅ )=0.367 sig(C ₆ )=0.439

sig(C₇)＝0.412sig(C₈)＝0.149sig(C₉)＝0.34sig(C₁₀)＝0.66sig(C ₇ )=0.412 sig(C ₈ )=0.149 sig(C ₉ )=0.34 sig(C ₁₀ )=0.66

再根据公式(3)计算权重:Then calculate the weight according to the formula (3):

w＝[0.260.210.2410.289]w₁＝[0.630.37]w₂＝[0.300.330.37]w = [0.260.210.2410.289]w ₁ =[0.630.37]w ₂ =[0.300.330.37]

w₃＝[0.4520.410.138]w₄＝[0.520.48]w ₃ =[0.4520.410.138] w ₄ =[0.520.48]

根据公式(2)、(3)计算出组合权重:Calculate the combination weight according to formulas (2) and (3):

$w w (({C C}_{11})) = = \frac{33}{2525} w w (({C C}_{22})) = = \frac{33}{2525} w w (({C C}_{33})) = = \frac{11}{2525} w w (({C C}_{44})) = = \frac{22}{2525} w w (({C C}_{55})) = = \frac{33}{2525}$

$w w (({C C}_{66})) = = \frac{33}{2525} w w (({C C}_{77})) = = \frac{33}{2525} w w (({C C}_{88})) = = \frac{11}{2525} w w (({C C}_{99})) = = \frac{33}{2525} w w (({C C}_{1010})) = = \frac{33}{2525}$

计算基于AHM的权重:Compute weights based on AHM:

单层指标权重：Single-layer index weight:

w＝[0.2510.2630.2440.242]w=[0.2510.2630.2440.242]

w₁＝[0.60.4]w₂＝[0.2510.3720.377]w₃＝[0.4910.3020.207]w₄＝[0.3920.608]w ₁ =[0.60.4]w ₂ =[0.2510.3720.377]w ₃ =[0.4910.3020.207]w ₄ =[0.3920.608]

组合权重：Combination weight:

$\overset{~ ~}{{w w}_{11}} = = [[\begin{matrix} 0.128 0.128 & 0.109 0.109 \end{matrix}]] \overset{~ ~}{{w w}_{22}} = = [[\begin{matrix} 0.076 0.076 & 0.083 0.083 & 0.092 0.092 \end{matrix}]] \overset{~ ~}{{w w}_{33}} = = [[\begin{matrix} 0.117 0.117 & 0.113 0.113 & 0.046 0.046 \end{matrix}]] \overset{~ ~}{{w w}_{44}} = = [[\begin{matrix} 0.107 0.107 & 0.129 0.129 \end{matrix}]]$

计算最终的权重Calculate the final weight

通过粗糙集与AHM分别获得主客观评价指标的权重，计算组合评价，根据公式W＝μw_Ai+(1-μ)w_Bi，取μ＝0.62，结果偏向客观权重，综合权重如表2、3所示。Obtain the weights of subjective and objective evaluation indicators through rough set and AHM respectively, and calculate the combined evaluation. According to the formula W=μw _Ai +(1-μ)w _Bi , take μ=0.62, and the result is biased toward objective weights. The comprehensive weights are shown in Tables 2 and 3 shown.

表2各二级指标的权重Table 2 The weight of each secondary index

表3各一级指标的权重Table 3 The weight of each first-level indicator

由此可以看出任务流程能效指标是机床产品制造系统能效评价的重要因素。It can be seen that the energy efficiency index of the task process is an important factor in the energy efficiency evaluation of the machine tool product manufacturing system.

一级指标权重：W＝[0.2570.2300.2420.271]Level 1 indicator weight: W＝[0.2570.2300.2420.271]

二级指标权重：W₁＝[0.620.38]W₂＝[0.2810.3460.373]Secondary index weight: W ₁ ＝[0.620.38]W ₂ ＝[0.2810.3460.373]

W₃＝[0.4670.3690.164]W₄＝[0.4710.529]W ₃ =[0.4670.3690.164]W ₄ =[0.4710.529]

定量指标的无量纲化处理Dimensionless Treatment of Quantitative Index

对甲机床厂的各指标经过无量纲化处理得出:After dimensionless processing of each index of A machine tool factory:

C＝[0.360.130.770.850.660.50.750.50.251]C=[0.360.130.770.850.660.50.750.50.251]

对乙机床厂的各指标经过无量纲化处理得出:After dimensionless processing of each index of B machine tool factory, it can be obtained:

C＝[0.580.640.620.0360.3310.240.910.5]C=[0.580.640.620.0360.3310.240.910.5]

确定定性指标的分值Determining Score Values for Qualitative Indicators

甲、乙机床厂的产品用能水平等级为“良”、“中”，对应的分值为3、2。The product energy consumption levels of A and B machine tool factories are "good" and "medium", and the corresponding scores are 3 and 2.

单指标模糊评价Single index fuzzy evaluation

确定各指标的隶属度，通过计算得：To determine the degree of membership of each index, through calculation:

R_i＝[0.50.60.60.70.60.70.40.50.40.5]R _i =[0.50.60.60.70.60.70.40.50.40.5]

一级灰色综合评价Level 1 Gray Comprehensive Evaluation

确定最优指标集C^*，并经过定量指标的无量纲化处理和定性指标的定量化处理：Determine the optimal index set C ^* , and go through the dimensionless processing of quantitative indexes and the quantitative processing of qualitative indexes:

C^*＝[0001111111]C ^* = [0001111111]

根据公式计算得：TOW_min＝0TOW_max＝0.964Calculated according to the formula: TOW _min = 0TOW _max = 0.964

取ρ＝0.5计算得：Take ρ=0.5 to calculate:

L₁(1)＝0.57L₁(2)＝0.79L₁(3)＝0.38L₁(4)＝0.76L₁(5)＝0.59L ₁ (1) = 0.57 L ₁ (2) = 0.79 L ₁ (3) = 0.38 L ₁ (4) = 0.76 L ₁ (5) = 0.59

L₁(6)＝0.49L₁(7)＝0.66L₁(8)＝0.49L₁(9)＝0.39L₁(10)＝1L1( ₆ )＝ _0.49L1 (7)＝ _0.66L1 (8)＝ _0.49L1 (9)＝0.39L1(10)＝ ₁

L₂(1)＝0.45L₂(2)＝0.43L₂(3)＝0.44L₂(4)＝0.33L₂(5)＝0.43L₂(6)＝1L₂(7)＝0.39L₂(8)＝0.83L2(1)＝ _0.45L2 ( ₂ )＝ _0.43L2 (3)＝ _0.44L2 (4)＝ _0.33L2 (5)＝ _0.43L2 ( ₆ )＝1L2(7)＝ _0.39L2 (8)＝0.83

L₂(9)＝1L₂(10)＝0.49L ₂ (9) = 1L ₂ (10) = 0.49

$R R = = [\begin{matrix} 0.57 0.57 & 0.45 0.45 \\ 0.79 0.79 & 0.43 0.43 \\ 0.38 0.38 & 0.44 0.44 \\ 0.76 0.76 & 0.33 0.33 \\ 0.59 0.59 & 0.43 0.43 \\ 0.49 0.49 & 11 \\ 0.66 0.66 & 0.39 0.39 \\ 0.49 0.49 & 0.83 0.83 \\ 0.39 0.39 & 11 \\ 11 & 0.49 0.49 \end{matrix}]$

根据公式J＝W×R得According to the formula J=W×R

${J J}_{11} = = [[\begin{matrix} 0.629 0.629 & 0.38 0.38 \end{matrix}]] \times \times [\begin{matrix} 0.57 0.57 & 0.45 0.45 \\ 0.79 0.79 & 0.43 0.43 \end{matrix}] = = [[\begin{matrix} 0.654 0.654 & 0.442 0.442 \end{matrix}]]$

${J J}_{22} = = [[\begin{matrix} 0.281 0.281 & 0.346 0.346 & 0.373 0.373 \end{matrix}]] \times \times [\begin{matrix} 0.38 0.38 & 0.44 0.44 \\ 0.76 0.76 & 0.33 0.33 \\ 0.59 0.59 & 0.43 0.43 \end{matrix}] = = [[\begin{matrix} 0.59 0.59 & 0.398 0.398 \end{matrix}]]$

${J J}_{33} = = [[\begin{matrix} 0.467 0.467 & 0.369 0.369 & 0.164 0.164 \end{matrix}]] \times \times [\begin{matrix} 0.49 0.49 & 11 \\ 0.66 0.66 & 0.39 0.39 \\ 0.49 0.49 & 0.83 0.83 \end{matrix}] = = [[\begin{matrix} 0.553 0.553 & 0.747 0.747 \end{matrix}]]$

${J J}_{44} = = [[\begin{matrix} 0.471 0.471 & 0.529 0.529 \end{matrix}]] \times \times [\begin{matrix} 0.39 0.39 & 11 \\ 11 & 0.49 0.49 \end{matrix}] = = [[\begin{matrix} 0.713 0.713 & 0.73 0.73 \end{matrix}]]$

二级灰色综合评价Level 2 Gray Comprehensive Evaluation

R＝[J₁J₂J₃J₄]^T，根据公式J＝W×R，R＝[J ₁ J ₂ J ₃ J ₄ ] ^T , according to the formula J＝W×R,

J＝[0.630.58]J＝[0.630.58]

根据计算分析可得，甲机床厂的企业能效为0.63，乙机床厂的企业能效为0.58。甲、乙机床厂的能效综合评价等级都属于中级,四个二级能效指标中产品能效指标起最主要的作用。乙机床厂的设备能效指标、任务层能效指标与甲机床厂相比差距较大，乙机床厂应在设备能效指标、任务层能效指标上加以重点改进，设备上需要更换或改进，任务流程上可能需要改进工艺路线或车间调度方法。According to calculation and analysis, the enterprise energy efficiency of A machine tool factory is 0.63, and the enterprise energy efficiency of B machine tool factory is 0.58. The energy efficiency comprehensive evaluation grades of A and B machine tool factories are all intermediate, and the product energy efficiency index plays the most important role among the four secondary energy efficiency indicators. Compared with A machine tool factory, the equipment energy efficiency index and task level energy efficiency index of B machine tool factory are quite different. B machine tool factory should focus on improving the equipment energy efficiency index and task layer energy efficiency index. The equipment needs to be replaced or improved, and the task flow Routing or shop scheduling methods may need to be improved.

本发明旨在为机床产品制造系统能效评价技术领域提供一种能效综合评价的方法,本领域的非普通技术人员在阅读本发明说明书的基础上可以对所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换而这些修改或者替换,并不使相应技术方案的本质脱离本发明各实施技术方案的精神和范围。The present invention aims to provide a method for comprehensive evaluation of energy efficiency in the technical field of energy efficiency evaluation of machine tool product manufacturing systems. Those skilled in the art can modify the technical solutions recorded on the basis of reading the description of the present invention, or modify them Some technical features are equivalently replaced, and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions implemented in the present invention.

Claims

1., based on the machine tool product manufacturing system efficiency evaluation method of grey fuzzy algorithm, it is characterized in that, comprise the following steps:

Step one, set up machine tool product manufacturing system efficiency assessment indicator system, in efficiency assessment indicator system, all specific targets form factor of evaluation collection C;

The weight set W of the combined method agriculture products of step 2, using rough collection and analytical hierarchy process; Namely utilize rough set and analytical hierarchy process to obtain the index weights of objective, subjective two aspects respectively, carry out comprehensively, obtaining last index weights to both, obtain one group of final evaluation criterion weight

W＝μw _Ai+(1-μ)w _Bi

Wherein w _airefer to objective weight value, w _birefer to subjective weighted value, μ ∈ [0,1], the value of μ is determined as the case may be, close to 0, μ more represents that decision-making more tends to expertise, close to 1, μ more represents that decision-making more tends to objective data;

The method of step 3, application linear scale transform carries out nondimensionalization process to the original quantitative target data of machine tool product manufacturing system;

Step 4, application classification scoring carry out quantification process to the original qualitative index data of machine tool product manufacturing system;

Step 5, application triangle are subordinate to model determination single factor test fuzzy evaluation collection;

Step 6, calculate first class index Evaluations matrix according to Grey Incidence, and then obtain first class index evaluation result;

Step 7, Grey Incidence comprehensive evaluation is utilized to go out multilayer index.

2. as claimed in claim 1 based on the machine tool product manufacturing system efficiency evaluation method of grey fuzzy algorithm, it is characterized in that, described in step one, efficiency assessment indicator system comprises economic energy efficiency indexes, product energy efficiency indexes, energy efficiency of equipment index and flow of task energy efficiency indexes 4 first class index, the two-level index that described economic energy efficiency indexes comprises has: ten thousand yuan of product energy consumptions, ten thousand yuan of added value energy consumptions, the two-level index that described product energy efficiency indexes comprises has: unit product comprehensive energy consumption, unit product amount of energy saving, product energy level, the two-level index that described energy efficiency of equipment index comprises has: machine tool efficiency, energy transfer efficiency, energy processing conversion efficiency, the two-level index that described flow of task energy efficiency indexes comprises has: production technology efficiency, resources of production scheduling efficiency, these 10 two-level index form factor of evaluation collection C.

3. as claimed in claim 1 based on the machine tool product manufacturing system efficiency evaluation method of grey fuzzy algorithm, it is characterized in that, in step 3, if the raw value of a kth index is then to carry out nondimensionalization process through following formula, the data value C wherein after process _i(k) ∈ (0,1),

C_{i} (k) = \frac{c_{k}^{i} - {minc}_{k}^{i}}{{maxc}_{k}^{i} - {minc}_{k}^{i}}

And i=1,2n, k=1,2m, wherein m is decision index system quantity, and n is possibility quantity.

4. as claimed in claim 1 based on the machine tool product manufacturing system efficiency evaluation method of grey fuzzy algorithm, it is characterized in that, step 4 is converted into quantitative target qualitative index, adopts classification scoring, gives a score value to every grade.

5. as claimed in claim 1 based on the machine tool product manufacturing system efficiency evaluation method of grey fuzzy algorithm, it is characterized in that, step 5, from single index, determines the degree of membership evaluating element of set element; The FUZZY MAPPING of (V) from U to F:

f : U &RightArrow; F (V), &ForAll; u_{i} &Element; U, u_{i} | &RightArrow; f (u_{i}) = \frac{r_{i, 1}}{c_{1}} + \frac{r_{i, 2}}{c_{2}} + ... + \frac{r_{i, k}}{c_{k}} ... + \frac{r_{i, m}}{c_{m}}

In formula, r _i,krepresent u _ibelong to c _kdegree of membership.

6., as claimed in claim 1 based on the machine tool product manufacturing system efficiency evaluation method of grey fuzzy algorithm, it is characterized in that, step 6 realizes the comprehensive evaluation of first class index according to Grey Incidence, and optimum index set is:

C * = [\begin{matrix} c_{1}^{*} & c_{2}^{*} & ... & c_{m}^{*} \end{matrix}],

Iotave evaluation matrix is:

D = [\begin{matrix} c_{1}^{*} & c_{2}^{*} & ... & c_{m}^{*} \\ c_{1}^{1} & c_{2}^{1} & ... & c_{m}^{1} \\ ... & ... & ... & ... \\ c_{1}^{n} & c_{2}^{n} & ... & c_{m}^{n} \end{matrix}]

In formula, m is decision index system quantity, and n is possibility quantity, for the optimal value of a kth index, it is the original value of a kth index in i-th scheme; The two poles of the earth lowest difference can be drawn:

The maximum difference in the two poles of the earth:

{TOW}_{m a x} = \underset{i}{m a x} \underset{k}{m a x} | c_{k}^{*} - c_{k}^{i} |

Grey incidence coefficient is:

\begin{matrix} L_{i}^{k} = \frac{{TOW}_{m i n} + {ρTOW}_{\max}}{| c_{k}^{*} - c_{k}^{i} | + {ρTOW}_{m a x}}, & ρ &Element; (0, 1) \end{matrix}

Evaluations matrix is:

R = [\begin{matrix} L_{1} (1) & L_{2} (1) & ... & L_{n} (1) \\ L_{1} (2) & L_{2} (2) & ... & L_{n} (2) \\ ... & ... & ... & ... \\ L_{1} (m) & L_{2} (m) & ... & L_{n} (m) \end{matrix}]

Last Grey Comprehensive Evaluation:

J＝W×R

In formula, W is weight matrix, and R is Evaluations matrix.

7., as claimed in claim 6 based on the machine tool product manufacturing system efficiency evaluation method of grey fuzzy algorithm, it is characterized in that, step 7 realizes multistage Grey Comprehensive Evaluation: if index has y layer, then will carry out y level Grey Comprehensive Evaluation, c _kas a kth evaluation index, its single index evaluation collection wherein s is as index quantity; When index have two-layer and every layer have multiple index time, first single index fuzzy evaluation is carried out to second layer index, by second layer index, one-level Grey Comprehensive Evaluation is carried out to ground floor index again, carry out secondary Grey Comprehensive Evaluation by the one-level Grey Comprehensive Evaluation result of ground floor index to second layer index again, evaluation result is system evaluation result.